DTIC AD0010203: THE "AVERAGE MAN" ?

UNCLASSIFIED

AD NUMBER

AD010203

NEW LIMITATION CHANGE
TO

Approved for public release, distribution
unlimited

FROM

Distribution authorized to U.S. Gov't,
agencies only; Administrative/Operational
Use; DEC 1952. Other requests shall be
referred to Aero Medical Center, Wright
Air Development Center, Wright -Patter son
AFB, OH 45433.

AUTHORITY

AMRL ltr, 25 Apr 1977

THIS PAGE IS UNCLASSIFIED

I ! < IlM* M. M» ! i VU HI) M.

f ». .
CD
CJ>

LH

CTI

oo

«=C <£

THE "AVERAGE MAN"?

oii.nr.kr s, oaxiei.s, i si i. r, vsap <msc)
M-llO Ml: 1)1 CAl. LAltORA TOW V

DEC EMU kit l<jr,2

NOTK.lv. ! )i >* tr iliti t •<> n of ilii>* report outside Government
will be mndc only with the prior upprovnl of the Atfo
Mcdic.il I .uborutorv, UAI)(.i.

1 RIGHT AIR UEVEIjOPMKNT CSNTF.R

technical note *-CRD yyl

THE '‘AVERAGE MAN'*?

Gilbert S. Daniels. 1*1 Lh W>iF (NSC)
Aero Nerii’ iil laboratory

December 1952

HDO No. 69S-71

bright Air Development Cesier
Air Research ■roil Development Communt!
Veiled ,MJte» Air For-e

F-ree Uui-> , Ohi >

r.

FOREWORD

Tills Technical Note was prepared from statistics compiled
by Antioch College, Yellow Springs, Ohio under AF Contract No.
18(600)-30. Research and Development Order No. 695-71, ,!USAF
Personnel Body Anthropometry, " is the applicable project and
1st Lt Gilbert S. Daniels, U5AF (M3C), the project scientist.

Sincere acknowledgment is made to Mr. Edmund Churchill of
Antioch College who compiled and prepared the appendix of this
report.

INTHGDUCTION

The tendency to think- in torms of the "average man*
ie a pitfall into which many pernons blunder when
attempting to apply human body size data to design
problems. Actually it ic virtually iispo&sible to find an
"average nan" in the Air Force population, This is not
becauuu of any unique traite of thia group ox men, but
because of the great variability of bodily dimensions
which ie characteristic of all men* It is the intent of
this Technical Hote to point out and explain bods of the
factors that lead to the difficulties arising from the use
of "average* dimensions and to indicate to seme extent how
they may be avoided*

The data on which this Technical Hote is based are
the results of the Air Force Anthropometric Survey of
1950. There is, however, every reason to suppose that
conclusions similar to th'se reported here would have been
reached If the same type of analysis had been applied to
body size data based on almost any group of people.

TK fcCKr-53-7

1

THE " AVERAGE MAS"

The "Average Man" is a very prominent figure, at - general . ule he
is used as an oversimplified means of describing the combined characteristics
of a varied population. Thus we are presented with an "average man* who
Is about 5 feet 9 inches tall although the population he represents may
vary from under 5 feet to over 6 feet 6 inches in stature. Clearly, if we
were to ise the 5 foot 8 inch dimension for the design of au opening such as
a doorway, we would have a door through which the "average man", and indeed
all of the people who are shorter than average, ccnld pae« ^ ' * ;ructod.

The taller poople, however, and these would represe*. . w aOuu 6 50/& o, > ~ ov*."1

population, would have to stoop to avoid hitting their neadr on top of the
doorway. Obvious as this example say be, it does illustrate that design
problems require the application of human dimensional data beyond the over-
simplified and Inadequate dimensions of the "average man".

Data published (1) by or available through the Anthropology Section of
the Aero Medical Laboratory make available the information needed by the
project engineer or scientist for the design of Items where human dimensions
are of importance. These data extend the description of a particular dimen-
sion beyond the average and thereby make it possible to select a suitable
range of the dimension as It occurs in the total population. Such o range
say, of course, be limited to serve the particular project in question.

Thus If staturo is being considered, a range of from 5 feet 5 Inches to 6
feet 1 Inch Is shown to cover 90$ of the Air force population, the range
having been trimmed to leave out the tallest $f> and shortest 5$ of the men.

If thM information Is used to design a doorway adequate for men 6 foot 1
inch In stature, 95$ of the total using group will be able to pass through
the doorway without hitting their heads.

While the use of average dimensions is generally unsatisfactory even
when only one dimension is being considered at a time, the Inadequacy of
the "average man" method it compounded many times when more than one dimen-
sion is to be considered in a design problem. As an abstract representation
of • mythical Individual most ropresentatlve of a given population, the
"average awn" is convenient to grasp in our minds. Unfortunately he doesn't
exist. Instead of being the easiest individual of a group to provide for,
and the most common, the "average man" is in reality a very rare specimen
and very hard to fit,

The fallacy of the "average man" concept is further illuctratod by a
stady based on body measurements made on over hfc^0 Air Fore*- flying
personnel. From a total of 13? available measurements a smaller group,
all useful in clothing design selected. (Measurements applicable to
other problems such as cockpit layout or seat design could equally veil
have boon chosen; thoy would luue given much t.io oano results). The records
of tho u.oOO-pJ’.ts nor. *«■.-(■ then examined to see how many of these men, if
(i i.y , could t-e consiacrod average in all of the selected measurement s.

The exact method of deriving thifl range Is discussed in detail in the
Anr’srdix hut for our inmodiafce purpose it is sufficient to state that
lncludee approximately the middle 3 0 % of the total population. This i
considerably more generous portion of the group than is included by th
average value. In a r attempt to find an "average man", the "approxima
average" ranges of each measurement were used as hurdles in a atep-by-
elimination. *h»n the full sample of 4063 men vae examined for statu
only 1055 fell within the acceptable range of "approximately averages
group of 1055 was then culled for all of the raen of "average" stature
also had "average’’ chest circumference. Only 302 of the group still q
Elimination was continued with additional "average8 dimensionc until a
end of 10 stops there was not a single individual remaining who fell v
the ’'average" range for all measurements. This process of elimination
gressod ca follows:

1. of tha original 4063 men

1055 were of approximately average stature

2. of these 1055 men

302 were also of approximately average chest- clrcuafi

3. of theae 302 r.ien

143 were aleo of approximately average sleeve length

4. of these l4;j men

73 ware also of approximately averagt crotch height

5. of these 73 men

28 were also of approximately average torso circmaf/

6. of these 28 man

12 were also of approximately average &i£ fl>r<rWfftr

7. of these 12 men

6 were alao of approximately average f&sk fii££iys££I

8* of these 6 men

3 were also of appr oiimatoly average jfifclet sJjC&HBtfll

9, of these 3 men

2 were also of approximately average jh&jgh

10. of these 2 men

0 were also of approximately average in crotch l.eite.tf

As a further step, TOO measurements made on one of the twn men remalr
after the ninth step were examined, Cf theco only 62 fell within tJ»*
of the middle 4u$ cf the total population; 29 were smaller (below tbe
percentile) end 9 wore decidedly high (above the ?0th percentile). 1
lOTgo group of measurements madi on the other of these two "most new
average" men showed much tho came pattern.

Th VCi'D-5

3

The exact method of deriving this range is discussed in detail in the
Aupendix hut for our iamediote purpose it is sufficient to state that it
includes approximately the Biddle 30% of the total population. This is a
considerably more generous portion of the group than is included by the exact
average value. In an attempt to find an "average Ban", the "approximately
average" ranges of each measurement were used as hurdles in a step-by-step
elimination. When the full sample of 4063 men vae examined for stature,
only 1055 fell within the acceptable range of "approximately average! This
group of 1055 was then culled for all of the "ien of "average" stature who
also had "average" chest circumference. Only 302 of the group still qualified.
Slim? nation was continued with additional "average* dimensions until at the
end of 10 steps there was not a single individual remaining who fell within
the "average" range for all measurements. This process of elimination pro-
gressed as follows:

1. of the original 4063 men

1053 were of approximately average staturo

2. of theee 1055

302 were also of approximately average chest- circumference

3. of these JOZ men

143 wars also of approximately average sleeve length

4. of these 143 men

73 were also of approximately average crotch height

5. of theee 73 men

28 were also of approximately average torso circumference

6. of these 28 man

12 were also of approximately average frip oircuaforance

?. of these 12 men

6 were also of approximately average neck circumference

8, of theee 6 men

3 were also of approximately average walBt circumference

9, of those 3 »en

2 were also of appr orimately average thicl^ circumference

10. of these 2 mon

0 were also of approximately average iu crotch length

As a further step, 100 measurements made or one of the two men remaining
after the ninth step were explained. Of theco only 62 fell within tl *- range
of the middle 4<$ of the total population: 29 were Biealler(bolcw the 30th
percentile) anu 9 vert, decidedly high (aoove the 70th percentile). A
iarc« c?o\\p of 00* euro monte mode on the other of then® two "most nearly
average" rnn ohevod vnch the same pattern.

Ti* WCH&-5V7

Thus it can be seen that the "average nan" is a misleading and illusory
concept as a basis for design criteria, and i3 particularly so vhon more than
one dimension is being considered.

This Technical Note is not meant to d. scourage the project scientist,
however, but merely to point out that the "average man" is usually not
the solution to the design problem. Data are available which give a far
more accurate description of range of human variability in body dimensions.
r report is in progress (2) which fully describes each of 131 separate
measurements of the Air Force flying population. While net yet available in
published form, the data of this report are available through the Anthropology
Section, Aero Medical Laboratory, Wright Air Development tenter, Wright-
Patterson Air Force Base, Ohio, for use by government agencies and authorized
contractors. In addition, the Section can make available dimensional design
criteria for cases where more than one dimension is being considered at a
time. Those multidimensional descriptions cannot be published in advance
since each cno is a separate problem in itself and the number of such
combinations of measurements approaches the infinite. For the benefit of
project engineers whose problems are in the mu tl dimensional category, custom-
designed solutions are available through the Anthropology Section, WCRDB-2,
(Telephone 2-3230) .

PUBLICATION REVIEW

This report has been review and is approved.
FOR. THE COMMANDING GENERAL:

£031

Colonel, USA? (MC)

Chief, Aero Medical Laboratory

Directorate of Research

nt vcsD-i j-''

4

API- SHI) LX

Method of Deriving "Approximate Average"

Any definition of the "approximate average" is necessarily arbitrary.

The statistical average of a measurement is so narrow in range that the-
oretically it includes only those Individuals whose measurement is exactly
the same as the mean value* In practice this is expanded to include those
individuals who fall within the same class interval as the mean value* Tor
instance, since st&turo is tabulated in 1 centimeter Intervale, all of those
individuals who are within plus and minus one half centimeter of the mean
value would be considered average* This range would Include less than 6,U$
of the total group and is far too restricting for actual application of
data* For the present purpose we have defined "average" wore liberally to
include nil individualo who fall within a range of plue or .sinus three-tenths
of a standard deviation of the mean value* This definition eeeme reasonable
on two counts:

1. It results in ths middle 25 or 3<9> of a group being classified
as approximately average on any one measurement.

2, It provides for moet measurements a value range which is at
least the equivalent of a full clothing else (over l£ Inches
on chest circumference, over 1 inch for sleeve length, etc.)

Actually it was necessary to use Halts which vers slightly different
and, in general, slightly wider than those just described becauoo of the
form In which the dmta were available* The limits were set to the nearest
whole value of measurement, i.e*, the nearost whole centimeter.

Table I gives, for each of the mensureaonto us ad in this study, the
values of the mean and the standard deviation, the range of values taken to
be approximately average, and the number of men who fell within this range*
Full descriptions of tbsse measurements and of the sen Included in the
sample population can bs found in WADC Technical Report 5?-321 "Anthropometry
of Flying Personnel-1950" (2).

TV V('Rb-ys-7

5

TABU! I

Haaeuroacnt

■Hgftfl

*N for Hau

ot-!Asaaa

Stature

175.5 ce

6.2 cm

173.95*177.95 ca

1055

Choet Circumference

98.6 ca

6.2 cm

96.95-100.95 ca

1023

Sleeve Length

85.5 ca

3.8 ca

33.95-86.95 ca

1253

Crotch Height

83.4 ca

4.4 ca

81.95-84.95 ca

1105

Vertical Trunk Ciro.

164.6 ca

7.3 ca

162.95-166.95 ca

1093

Hip Clrcuaference (Sit.)

106.0 ca

7.2 ca

103.95-108.95 ca

1113

Heck Clrcuaference

38.0 ca

1.9 ca

36.95- 38.95 ca

1713

Valet Clrcuaference

81.4 ca

7.7 ca

78.95- 83.95 ca

990

Thigh Clrcuaference

56.9 ca

4,4 ca

54.95- 57.95 c«

1114

Crotch Length

71.6 ca

5.1 ca

69.95- 72.95 c«

1008

• B*««4 oo the original croup of 4,063 aen.

71 V CCD-53 -7

b

BIBLIOGRAPHY

AF Technical Report 5501 Hunan Body Size in Military,. Aircraft and
Personal Equipment. AF, AM 3, Wright Field, Dayton, Ohio (Unclassified)

WADC Technical Report 52-321, Anthropometry of Flying Personnel. 1950,
Wright Air Development Center, Wright-Patterson Air Force Base, Ohio.
(Unclassified) (In preparation)